AiMesh Node going OFFLINE after a few weeks

[OzarkEdge]

Now be fair, I've been using PuTTY for the thick end of two decades and they only shipped a built-in SSH client in 2018, if 10 seconds of googling got that right. (Otherwise, absolutely, don't install an extra program if there's a decent built-in solution!)

I share how I do things to help others, not to be unfair. I've only had Win10 since 2020 so it was only natural to look under the hood for an SSH client.

OE

 

[Dutch Schaefer]

I have a Gt-AX11000 as main router and the RT-AX88u as node and i get the Zombie treatment by the node


the RT-AX88u goes Zombie or offline if you prefer after a day or 2... it seems it's still connected to the network since i can ping the rooter, but i suspect the wireless isn't working anymore on the node, ehternet LAN is still working on it.

it's still accessible via the APP (the 88u node) but i don't see the node in the GT-AX11000 web interface or APP.


I disconnect the 88u and it is staying in zombie mode after the reboot.

I haven't try to reboot the main router yet(Gt-AX11000) but after reading previous posts here, i suspect it won't work either...

i use ethernet backhaul with 2 different SSID for 2.4 and 5 and i have the latest asus firmwares 3.0.0.4.386_41700


can somone share info on how to setup the node in AP (access point) mode instead of AIMESH for the best results since AIMESH seems to be a AImess ?
Like the settings to optimize sharing between the 2 rooter for bands(20/40/80) and canals(100 to 165) and others settings that need to be touched)


Help is appreciated (sorry for english) it's not my first language

 

[Thw0rted]

The initial setup wizard should let you pick AP mode instead of AiMesh, if you really want, but you shouldn't have to do that. Have you tried a full factory reset of both devices, set up from scratch? I know it's a pain but it's the first thing everybody on these boards is going to ask you to do.

 

[Dutch Schaefer]

it's the second time i'm doing it, the setup from scratch and your right it's a bit of a pain since it takes 10minutes each time. Yes it works when i'm doing it but it seems to crash after a few days and the zombie node appear again.

 

[Thw0rted]

It's interesting that it shows up in the app. So, to be clear:

* you see the zombie in the app
* you don't see it in the web interface ("AiMesh" page)? Or it's listed but greyed out / disabled / offline?
* Devices plugged into the zombie are fine
* Devices cannot connect to the zombie via WiFi

I'm just curious because I've had this behavior before. At the time, the zombie node had the DHCP server for the house wired into it. I've since moved the DHCP server to hang directly off the router, and haven't had that particular problem since. (It might have been a slightly different problem, since power-cycling the zombie would usually bring it back online for a while.)

 

[Dutch Schaefer]

1-Yes i see it in the app (i choose the node(88AX) and i can enter the app) no wireless client are connected the node and i see the clients of the main rooter (gt-ax10000) but they are all listed as ethernet
2-Yes i don't see the node (ax88u) in the AI MESH page of the main router (GT-AX11000) , i don't see it , so there is no grey out or disabled
3- YES in fact i just realized the that the device (A PC) which is plugged in the NODE awakes the NODE when it's in use, the computer that was plugged on the node was close yesturday, i just opened it to check that the ethernet connection was still working, and now the node isnt zombie anymore and i see it in AI MESH from the GT-11000.... weird but i can confirm the WIRELESS was offline when it was in zombie mode. Now the wireless is working again... i can confirm the wireless of that node isn't often used in my house... maybe it goes in sleep mode (Zombie) when there is no wireless clients after a certain amount of time ? Wireless clients cannot wake up the node...
4-YES when the node is in zombie or sleep, the is no wifi on it, i can confirm that because the level of the signal when i'm next to the rooter compare when it's online

 

[ForkWNY]

I read through this whole thread tonight only because I've seen this "zombie/offline" AiMesh node problem (on 68U and 86U nodes) in a few cases...all cases were wired ethernet backhaul.

First case...86U AP router and 68U ethernet AiMesh node. 68U would go "offline" after an arbitrary amount of time, usually 5-7 days. After inspecting things, I found the problem to be the patch cable. It was a DIY cat6 cable that wasn't quite crimped properly. Replaced cable. Problem has not reoccurred (after 5-6 months).

Second case...multiple 68U nodes, all wired. One of the 68U's would drop "offline" randomly, and require a factory reset to bring back online and re-join as an AiMesh node. Long story short, this one took some time, but I eventually found a bent pin in the e-net jack that connected the node back to the main AP. Replaced the e-net jack, reset the 68U after updating its firmware, hasn't had any problems for many months.

Third case was a defective 68U that wouldn't operate reliably in any mode, as an AP, mesh node, etc. Tossed into trash can.

At any rate, I've never experienced "offline" zombies on my own network (5 AiMesh nodes), but in cases where I had observed this problem in 2 out of 3 cases with other users, it was related to a bad connection, either patch cable or jack itself. Sometimes these things are easily overlooked, my $0.02.

 

[Thw0rted]

Just to add a data point, my "zombie node" problem could also have been a bad cable -- I did move the DHCP server but the *reason* I moved it is because I wanted to switch from wired to wireless backhaul, because I knew the wired backhaul was dodgy. (You can't have your DHCP server attached to a wirelessly-connected node, because they did a bad job handling static IPs.)

 

[SignedAdam]

I'm happy to report since aimesh 2.0 I've had no nodes going zombie, I'm still using the power lines in my setup between the main node (gt-ax11000) and the node (rt-ac86u) using the same firmware as mentioned before, (3.0.0.4.386 on both models) no issues to speak of I'm very happy with the latest release

 

[Dutch Schaefer]

Just to add a data point, my "zombie node" problem could also have been a bad cable -- I did move the DHCP server but the *reason* I moved it is because I wanted to switch from wired to wireless backhaul, because I knew the wired backhaul was dodgy. (You can't have your DHCP server attached to a wirelessly-connected node, because they did a bad job handling static IPs.)

in my case it's not a bad cable

it's going into sleep mode (i don't know why), only way to wake the node up is to add another router or PC on a lan port of the NODE and it'a appearing instantanly in the AI MESH menu of the main router.

Node isn't a zombie anymore...

it's weird... I hope this will get corrected in next firmware, i've tried to plug a NAS and a phillips hue hub on a lan port of the node to see if it could awaken the node with no success, only a PC or another router are able to bring the zombie to life ....

 

[SignedAdam]

in my case it's not a bad cable

it's going into sleep mode (i don't know why), only way to wake the node up is to add another router or PC on a lan port of the NODE and it'a appearing instantanly in the AI MESH menu of the main router.

Node isn't a zombie anymore...

it's weird... I hope this will get corrected in next firmware, i've tried to plug a NAS and a phillips hue hub on a lan port of the node to see if it could awaken the node with no success, only a PC or another router are able to bring the zombie to life ....

Powerlines (the equipment that allows you to sent a ethernet signal through your house plug sockets) some of them have a power saving functionality, where if they aren't used, the powerlines switch off, if it's tp-link you have to download a program in order to change the options, some even allow you to update the firmware.

 

[Pergola Fabio]

Hi @Daddio or someone else
i have since a few months also 3 aimesh routers, main router rt-ac88u, with 2 nodes rt-ac86u, all running merlin 386.2.6 , all ethernet connected
after indeed a few weeks, i can see the nodes offline, allthough they are still network connected, i can still ping them, and also the devices that are connected with cable on the nodes, are still working
but wifi is just dead, and they show offline
a quick reset/restart with the button on the node, and all is fine again for some weeks

when it happened this morning, i see the node offline, and i see the upling type = wireless 5gh, thats verry strange, since i use ethernet
after the restart its normal again
so why is LAN dropped while it still works?

After reset :

priority is also 1 G wan, seems i am unable to change that setting? cant i specify like LAN ONLY?

i think it happened this night, if i look in log, its flooded with these :

Aug 6 01:17:06 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:17:38 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:17:38 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:18:11 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:18:11 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:18:43 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:18:43 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:19:15 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:19:15 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:19:48 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:19:48 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:20:20 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:20:20 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:20:53 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:20:53 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1
Aug 6 01:21:25 dnsmasq-dhcp[1969]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Aug 6 01:21:25 dnsmasq-dhcp[1969]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10 rt-ac-86u1

thnx in advance

 

[Adooni]

you are lucky AiMesh working for 3 weeks - a lot people have challenge to get it working for 1day w/o issues

therefore you can use (Administration > System > Enable reboot scheduler) to setup reboot.

 

[Pergola Fabio]

no, its quite stable , i need to wait for the next culprit , probably within a week, hopefully not, since i also upgraded to last merlin

 

[Pergola Fabio]

crap, spoke too soon, today a mesh was offline!!
i made somes screenshots, i am not home, so i access VPN to main router
the mesh = 192.168.0.2 .... still pingable!!!
1 wifi client (on mesh)=> dropped of
1 lan client (on mesh) => 192.168.0.41 => still pingable!!

see screens below
but it says that the mesh is disconnect... aslo management tools are unavailable
so i have no idea how i can reboot it from remote?? Maybe it helps that i reboot the main router?

why is my main roter saying its offline, while its perfectly reachable??

 

[OzarkEdge]

crap, spoke too soon, today a mesh was offline!!
i made somes screenshots, i am not home, so i access VPN to main router
the mesh = 192.168.0.2 .... still pingable!!!
1 wifi client (on mesh)=> dropped of
1 lan client (on mesh) => 192.168.0.41 => still pingable!!

see screens below
but it says that the mesh is disconnect... aslo management tools are unavailable
so i have no idea how i can reboot it from remote?? Maybe it helps that i reboot the main router?

why is my main roter saying its offline, while its perfectly reachable??

The AC88U LAN ports are known to die... perhaps one is affecting your backhaul comms. Or a dodgy cable.

The AC86U would make a stronger main router... and it has hardware encryption for faster VPN use. I would rebuild the network with the AC88U as a remote node.

Did you reset all firmware to factory defaults before configuring the router and adding the reset nodes?

OE

 

[Pergola Fabio]

I am only using the 4 first ports of the 88u, and they are not died...

My 2 nodes are indeed connected to them, and i can ping the node when it's offline...

I can even ping the devices connected to the node that was offline... Everything looks good... It's just that the node is showing "offline" and all wifi clients connected to it are dropped...

I can still do a ssh session to the node, if i then do a reboot command, a min later everything is back to normal

 

[OzarkEdge]

I am only using the 4 first ports of the 88u, and they are not died...

My 2 nodes are indeed connected to them, and i can ping the node when it's offline...

I can even ping the devices connected to the node that was offline... Everything looks good... It's just that the node is showing "offline" and all wifi clients connected to it are dropped...

I can still do a ssh session to the node, if i then do a reboot command, a min later everything is back to normal

So, you don't even want to try a different cable?

Did you reset all firmware to factory defaults before configuring the router and adding the reset nodes?

OE

 

[Pergola Fabio]

I tested the cables , all pins are correct... Otherwise I would not be able to ping it , i can even ping the devices connected to the node , when my desktop is connect to lan on another node ...

Everything is pingable, everything is reachable, it's just my main router that's complaining that's offline... So i just initiate a reboot command, and all back to normal ,for about 3 weeks...

I can still see in syslog, that the node is requesting every few sec a DHCP address from main router...

Yes , everything was factory resetted when I started to use the nodes, they are quite new , bought it like a half year ago...
At first, they were just configured as standalone AP, never had issues then..

It's when I configured them as nodes, the issues began

 

[Pergola Fabio]

Again, offline , if i do the command below on main router ,i still see it listed ... So why is it offline ?? It happened again after about a week uptime


nvram get cfg_device_list

@LimJK ..

Have you resolved it? I see exactly the same behaviour... When it's offline, i see also indeed a lot of DHCP request every 30 sec for that node... I can still ping it, i can still ssh it... Everything works like it should , but seems it's just offline state for the main router... And wifi is complete gone on that node... While ethernet works... Looks like its locked..
Always happens after about a week ...

Only thing to resolve it , is to reboot the node